Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

Department of Geography

Staff Profile

Publication details for Professor Louise Bracken

Rollason E., Bracken, LJ., Hardy R.J & Large A.R.G. The importance of volunteered geographic information for the validation of flood inundation models. Journal of Hydrology. 2018;562:267-280.

Author(s) from Durham

Abstract

Two dimensional flood inundation models capable of simulating complex spatially and temporally differentiated floodplain flows are routinely used to model and predict flooding. However, advances in modelling techniques have not been matched by improvements in model validation. Validation of flood models remains challenging due to a lack of available spatially-explicit data; traditionally measured data and validation approaches reveal little about the ability of a model to simulate the complex dynamics of floodplain flows, including the pathways, timeline, and impacts of an event. In this paper we propose a novel method for the validation of hydraulic models of flooding using quantitative and qualitative Volunteered Geographic Information (VGI). This method uses VGI data to enhance traditionally measured validation data by reconstructing the observed dynamics of a flood, allowing validation of the temporal and spatial simulation of these dynamics. We illustrate the method using a case study from Corbridge in the northeast of England, using VGI collected through participatory research with people affected by severe flooding in 2015. The results of the study demonstrate that VGI data can be used for the effective reconstruction of flood event dynamics. The results also reveal that the proposed validation approach is able to identify underperformance in the model’s simulation of event dynamics not evaluated by standard global performance measures. Such a lack of evaluation can have adverse consequences where dynamic model outputs are used locally to influence floodplain management. As a result, we propose a new framework for model validation, adopting a pragmatic and flexible approach to examining event dynamics using a diverse range of data.